TOPIC:

Hi, I just got my UMIK-1 a few minutes ago and went to download the calibration file. As I looked at it I noticed some pretty "big" correction values (+2.5 to +5dBs values all the way down from around 14kHz down to exactly 812Hz and then a sudden big drop at the next incremental measureing freq of 802Hz where it registers nearly perfect all the way to 20.6Hz where it begins to climb, this is acceptable I guess but then the correction file just ends at 20.1Hz ...huh??? Looking at the spec of this mic it says 15Hz-20kHz... so what's the deal with my mic's correction factors between 15-20Hz? Was calibration below 20Hz halted? The correction factor just starts a climbing trend at the last few measurements in the calibration file and so I don't have clue to what happens below 20.1Hz... does it suddenly get worse the way it does between 802Hz and 812Hz and above? This is very unfortunate as I consider this range particularly critical for my measurements as I purposely purchased the mic for accurate measurements of my sub down to at least the 15Hz it was advertised to be accurate to. I need some one to please explain this shortfall to me as right now I don't feel I can trust its measurements to be corrected properly.

What's your serial number? (I'm just curious to see the file.)

700-0275
My file is not as bad as some others I've viewed but there are many that are much better. It appears that about half of all the calibration files (I've looked at about 30 or so now) show a large jump going from 812Hz to 802Hz and that puzzles me. Either files have a nearly indiscernable difference passing through those frequencies or a sudden large jump when passing through those two frequencies but nothing really inbetween, which I find curious... could it be that two different calibration 'test beds' are being used and one has an anomaly having nothing to do with the calibration? In which case, the calibration file would be suspect, but which half... 812Hz and above or 802Hz and below? I just wish I'd gotten one without that large step at 812/802Hz.

Hopefully I'll get an offical (DevTeam) response to simply explain what's up with that 812/802Hz jump and also hopefully I'll also get an answer as to why the calibration file stops at 20Hz but specs say down to 15Hz. I'm hoping its just something I don't yet understand about the calibration process but I'll have to wait until a DevTeam member gets around to offering up some kind of response in this thread.

The one I have here is 700-0144. It looks like the calibrator was changed at number 700-0190, the file formats and frequency spacing change there.

Are you able to run a nearfield measurement of a driver?

I can but haven't. What do you suggest? I have a calibrated EMM-6 I could measure against. So far the only files I looked at were sequential around my file number (latest batch numbers)... didn't occur to me to try looking that far back (into the first batches). Hopefully there is some simple explanation to solve the mystery...

Excellent! If you can position the mic close to a woofer cone - say 5 cm - then you should be able to compare the mics directly (each with their respective cal file). Levels will be different but if there's a 2.5 dB drop just above 800 Hz in the UMIK measurement it would indicate an issue with the cal file.

It looks like the new calibration files have very large compensation factors from 800Hz and up.
The files before 700-0190 also have three columns instead of two.
Calibration files before 700-0190 is quite linear.
I will receive my mic any day now. I hope this problem can be fixed without sending the microphone back.

Same, it would be good to get some feedback on this.

monomer wrote:

I have a calibrated EMM-6 I could measure against.

Parts-express calibration, or an external one? (like cross spectrum)
It has been my experience that parts-express calibration file was totally meaningless...

You will have to provide me more than just a Tweet's worth of info here... you will have to actually define "meaningless" for me otherwise your posting will be meaningless. Yes, it is the Parts Express calibration file one downloads after providing the Serial Number... similar to the process one does for the UMIK-1 here.

I won't have the time or opportunity to compare the results between the two until Wednesday... but if the Parts Express calibration file is "meaningless" then I guess my only recourse would be to also include my RS digital SPL meter into the mix and they to see "which one's not like the others".

I got my emm6 through cross spectrum, who did the calibration file.
I asked them to select a very linear mic among the ones they had, because I wanted to use it with my old smaart live 5 that does not handle calibration files. I waited nearly 3 months to get that "special" one, much more linear than the average emm6 (first batches were even more linear it seems...)

I was able to compare their calibration to the one I downloaded from parts express (which is also provided in printed version in the box).

Here is the comparison:


The measurements and comparisons I did tend to prove cross spectrum's calibration is accurate. Which in turn tends to prove Dayton's one is... meaningless.

Here are some measurements of a random system, just to show the differences you get on a real worl measurement (please keep in mind that this particular emm6 was selected for linearity from the start)

I have extracted two frequencies(802.424 and 812.413) from several calibration files.
This must be a problem in the calibration rig or the microphones of the latest batch are really bad.

File 7000274.txt:
802.424 -0.0022
812.413 -3.4876

File 7000275.txt:
802.424 -0.0073
812.413 2.6590

File 7000277.txt:
802.424 0.0005
812.413 -4.0381

File 7000278.txt:
802.424 0.0068
812.413 -1.5635

File 7000280.txt:
802.424 0.0094
812.413 -2.2300

File 7000281.txt:
802.424 -0.0047
812.413 3.0787

File 7000282.txt:
802.424 -0.0016
812.413 2.7376

File 7000283.txt:
802.424 -0.0043
812.413 -1.9013

I have been advised that my UMIC-1 is sitting in Customs so I expect to receive the mic shortly. I also am concerned about the quality of the Calibration File and share the concerns already mentioned. Can the Dev team give us a status update on their progress in resolving this issue and if this issue can be resolved without returning my UMIC-1?

JohanW wrote:

I have extracted two frequencies(802.424 and 812.413) from several calibration files.
This must be a problem in the calibration rig or the microphones of the latest batch are really bad.

(...)

Maybe this gap is due to some mismatch between two measurement procedures used to calibrate the mic: pressure measurement for the LF, and normal gated measurement for HF. 800Hz would be a logical crossover frequency between those two techniques.
If this is the case this would be easy enough to resolve by just aligning the 800Hz and 812Hz values, and add the difference to subsequent values...

monomer wrote:

I won't have the time or opportunity to compare the results between the two until Wednesday... but if the Parts Express calibration file is "meaningless" then I guess my only recourse would be to also include my RS digital SPL meter into the mix and they to see "which one's not like the others".

The PE cal file won't vary wildly around 800 Hz, so it will be fine for this purpose. Actually, even just running a sweep with the UMIK will be enough to see if the cal file is correct. If anyone with one of these (post-0190 serial) would care to run a sweep that would help.

On the off chance someone reading this is in Sydney AU and would loan me their mic for a couple of days I can do the comparison.